The study objective is to assess the merits of the "white flight" thesis that holds that natives leave labor market areas that receive large numbers of immigrants because of the perceived economic and social costs associated with recent immigration. The migratory responses of native-born and foreign-born adults in the 1985-90 and 1995-2000 periods are examined to shed light on this thesis. The study premise is that if the foreign-born adults leave labor market areas of high recent immigration at rates comparable to native born or at higher rates, then that finding implies that forces other than recent immigration underlie internal migration patterns. The health implications of the work stems from the deleterious health conditions associated with concentration of ethnic minorities in poor, segregated communities. Since most immigrants initially settle in such communities, if they remain in them rather than moving to communities that offer improved health and social services, then morbidity/mortality differentials are likely to remain or even increase in the years ahead. 1990 and 2000 census data will be used to analyze out-migration from metropolitan labor market regions (LMRs) based on PUMS-L and PUMS-A 5% PUM. Both PUMS files use counties or county equivalents as building blocks. Contextual data on economic conditions and race-ethnic composition of the LMRs are appended to individual records drawing on county and metropolitan data sources. The Specific Aims are to assess whether: (1) there is a difference between foreign-born and native-born adults in the likelihood of out-migration from labor market areas of high immigration in the 1985-90 and 1995-2000 periods; (2) the migratory response patterns of foreign-born and native-born adults differs by gender; (3) the migratory responses differ depending upon the regional origins of recent immigrants; (4) native-born, non-Hispanic whites are more likely than native-born blacks, Hispanics, or Asians to show a migratory response to high immigration; and (5) the migratory responses of the foreign-born from different world regions of origin are lower if recent immigrants also come from the same region of origin. Logistic regression will be used to estimate the effects of individual characteristics, recent immigration, and LMR economic and ethnic-race context.